Pharmacokinetic (PK) Models for Chemical Risk Assessment of Perfluorooctanoic Acid (PFOA) and Perfluorosulfonic Acid (PFOS)

Dozens of PK models have been developed for PFOA and PFOS, which are ubiquitous environmental contaminants that accumulate in humans due to slow excretion. These models are essential for chemical risk assessment because they allow for interspecies dose extrapolations and derivation of human equivalent doses (HEDs). A review of recent PK modeling efforts was conducted to determine a suitable approach for HED estimation. A concurrent review of toxicology studies (not discussed here) showed rat, mouse, and monkey as the primary species needed for dosimetry modeling, with both adult and developmental studies informing critical toxicology endpoints for PFOA and PFOS. Hence, internal dose metrics from both adult and developmental animal studies needed to be translated to HEDs. For adult animal to adult human extrapolation, a simple algebraic calculation is used, since adult PFOA and PFOS concentrations are expected to be at steady state due to the relatively constant exposure over a period of many half-lives. However, concentrations in children and neonates are not at steady-state due to the relatively large changes in exposure coupled with rapid growth during development. Therefore, we implemented life-course pharmacokinetic models to predict animal and human serum concentrations across lifestages. The models were designed to predict serum concentration, as serum was the primary exposure measure in epidemiological studies and few animal outcomes had mechanistic information suggesting that a specific tissue concentration would better correlate with quantitative measures of observed toxic effect. The animal model describes pharmacokinetics in an individual using a single compartment with saturable resorption in the adult kidney, which leads to nonlinearity in the exposure-dose relationship. The models showed that humans excrete PFOA and PFOS more slowly than animals, resulting in a large discrepancy in predicted internal dose. For example, one study resulted in a 140-fold decrease in the predicted HED compared to the administered dose in the male rat. Comparatively, developmental scenarios demonstrated a similar pattern, with one study resulting in an HED 208 times smaller than the dose administered to mice dams. Finally, life-course human PK modeling was used to perform reverse dosimetry to interpret epidemiological studies relating adverse outcomes to measured serum concentrations.

Impact/Purpose

For the an updated chemical risk assessment on PFOA and PFOS, we performed a review of recent pharmacokinetic models and selected an approach for the extrapolation of animal doses and points of departures to human equivalent doses. The review, the implementation/modification of the selected models, and what we learned from the whole process will be presented at the upcoming meeting.

Citation

Zurlinden, T., P. Schlosser, B. Jacobs, V. Morozov, AND M. Dzierlenga. Pharmacokinetic (PK) Models for Chemical Risk Assessment of Perfluorooctanoic Acid (PFOA) and Perfluorosulfonic Acid (PFOS). Society of Toxicology Annual Meeting, San Diego, CA, March 27 - 31, 2022.